Part 1 — Eleven myths worth dismantling

Myth 1: "Quantum computers are exponentially faster than classical computers."

• The reality. Quantum computers are exponentially faster for a small set of specific problem structures — integer factorisation, certain simulation problems, some structured search. For everything else (spreadsheets, databases, web servers, training a large language model, running your ERP) they are slower, more expensive, and more error-prone than a classical laptop, and will be for the foreseeable future.

• Why it matters. Executives who hear "exponentially faster" assume quantum will replace or accelerate general computing. It will not. The business question is not "where do I use quantum instead of classical?" but "do my business problems match one of the narrow structures where quantum helps?"

Myth 2: "Quantum computers will replace classical computers."

• The reality. Quantum machines are specialised co-processors, analogous to GPUs. They will live inside heterogeneous systems alongside CPUs, GPUs, and classical HPC — handling specific sub-problems where they have an advantage and returning results to classical machines for everything else. Every serious quantum application today is a hybrid workflow. This will remain true even after fault tolerance arrives.

• Why it matters. This shapes architecture, procurement, and skills planning. A quantum roadmap that does not integrate with the classical HPC, cloud, and data strategy is a roadmap that will not deliver value.

Myth 3: "Quantum AI will transform machine learning."

• The reality. After a decade of quantum machine-learning research, no quantum algorithm has demonstrated convincing advantage over classical ML on a commercially relevant problem. Several early claims of exponential quantum speedup have been "dequantised" — reproduced classically at comparable cost. The most active current work is on quantum-enhanced generative models and quantum kernels for specific data structures. These are research directions, not business-ready technologies.

• Why it matters. "Quantum AI" is the single most over-sold phrase in the industry. If a vendor is leading with it, apply extra scrutiny. For companies already strong in classical AI, a small monitoring investment in QML is reasonable. For everyone else, it is a distraction.

Myth 4: "More qubits equal more powerful."

• The reality. Qubit count headlines are routinely misleading. A machine with 1,200 noisy qubits may be less useful than one with 20 very high-quality qubits for a given problem. What matters is the combination of count, gate fidelity, connectivity between qubits, coherence time, and the depth of circuits you can run before errors overwhelm the signal. A single meaningful metric — quantum volume, or the related algorithmic qubits — captures this better than raw qubit count, though neither is perfect.

• Why it matters. Vendor comparisons based purely on qubit count are close to useless. If a pitch emphasises qubit count over fidelity, connectivity, and error rates, that is a signal to ask harder questions.

Myth 5: "Quantum will break all encryption tomorrow."

• The reality. Credible timelines for a cryptographically-relevant quantum computer (one capable of running Shor's algorithm against RSA-2048) range from 2030 to 2040. No one knows the exact year. But this uncertainty is not reassurance: the migration to post-quantum cryptography takes 5–10 years for most enterprises, and adversaries are already harvesting encrypted traffic today to decrypt later. For long-lived sensitive data, the clock is already running.

• Why it matters. The risk is real and urgent, but the urgency is about your migration timeline, not about imminent decryption. Organisations that conclude "the quantum computer isn't here yet, so we can wait" are misreading the threat. Organisations that panic and deploy non-standard cryptography are also misreading it.

Myth 6: "QKD is the answer to the quantum threat to cryptography."

• The reality. It is not. Post-quantum cryptography (PQC) is the mainline global response to the quantum threat, endorsed by NIST, ENISA, NCSC, NSA, ANSSI, BSI, and essentially every major cybersecurity authority. Quantum Key Distribution is a specialised physical-layer technology with narrow but real applications — primarily in government, defence, sovereign financial infrastructure, and specific critical-infrastructure links. The NSA and NCSC have both explicitly stated that they do not recommend QKD for general commercial or national-security use.

• Why it matters. Vendors selling QKD as a general-purpose response to quantum risk are either confused or selling. For almost every business, the right investment is in PQC migration, not QKD.

Myth 7: "Quantum supremacy means commercial advantage."

• The reality. "Quantum supremacy" (or, more recently, "quantum advantage") refers to a quantum computer performing some computation faster than the best classical computer — but the computation is usually artificial, chosen to favour the quantum machine, and of no commercial value. Google's 2019 supremacy result and subsequent experiments are scientifically important milestones. They are not commercial products.

• Why it matters. A press release announcing "quantum advantage" tells you almost nothing about whether any commercial problem has become solvable. Commercial advantage — a quantum result that is the best available answer to a problem a paying customer actually cares about — has not yet been achieved. When it is, it will most likely come from quantum simulation, not from supremacy-style benchmarks.

Myth 8: "Quantum technology is just around the corner" / "Quantum is decades away."

• The reality. Both extremes are wrong. Quantum sensing is already generating real ROI in specific industries today. Quantum-safe security is a live, now-or-soon migration programme for every serious organisation. Quantum simulation is likely to deliver first commercial advantage this decade, in chemistry and materials. Fault-tolerant general-purpose quantum computing is 5–15 years away. Full quantum internet is 15+ years away.

• Why it matters. "Quantum is decades away" produces strategic inaction across the pillars where it is already relevant. "Quantum is here now" produces over-investment in pillars that are still research. The correct posture varies by capability, sector, and use case — which is exactly why a one-line answer to "when will quantum matter?" is always wrong.

Myth 9: "Entanglement allows faster-than-light communication."

• The reality. It does not. No information travels through entanglement; a classical channel is always required to compare measurement results and extract any usable signal. This is a popular-science misconception that sometimes leaks into business conversations and can undermine credibility.

• Why it matters. If a vendor or pitch implies otherwise, close the tab. Whoever wrote the material either does not understand the physics or is deliberately exploiting those who do not.

Myth 10: "Quantum technology is a single field."

• The reality. The word "quantum" in commercial use refers to five distinct technology families — computing, secure communications, cryptography-replacement, sensing, and simulation — with different physics, different maturity levels, different timelines, different vendor ecosystems, different regulatory environments, and different relevance to your business. Speaking about "the quantum market" or "the quantum roadmap" as a single thing is as useful as speaking about "the electronics industry" as a single thing.

• Why it matters. Strategy, budget, and procurement should be set per-pillar. Lumping everything into one "quantum programme" produces incoherent investment.

Myth 11: "To use quantum, you have to buy a quantum computer."

• The reality. Essentially nobody needs to. Quantum computers, simulators, and key-distribution systems are accessed via cloud APIs from hyperscalers (AWS Braket, Azure Quantum, IBM Quantum, Google Cloud) or directly from vendors. Quantum sensors are the one category you do procure as hardware — but even there, it is ordinary instrument purchasing, not cryogenic infrastructure. No enterprise we have seen needs to own a superconducting quantum computer in its datacentre. The marginal cases are national laboratories, hyperscalers, and a small set of defence operators.

• Why it matters. Any proposal that starts with "first we install the hardware" is almost certainly wrong for a commercial client. Cloud access makes early engagement cheaper and more flexible than hardware ownership.

Bonus myth: "Post-quantum means quantum."

• The reality. Post-quantum cryptography is classical cryptography designed to resist attack by a future quantum computer. It runs on ordinary hardware, uses ordinary software stacks, and has nothing to do with quantum physics in its operation. The "quantum" in the name refers to the threat model, not the technology.

• Why it matters. This confusion affects budget conversations (PQC is software work, not quantum hardware) and procurement (your cryptography vendor, not a quantum vendor, delivers PQC).